{"data":{"slug":"arize-ai-openinference","name":"openinference","tagline":"OpenTelemetry Instrumentation for AI Observability","github_url":"https://github.com/Arize-ai/openinference","owner":"Arize-ai","repo":"openinference","owner_avatar_url":"https://avatars.githubusercontent.com/u/59858760?v=4","primary_language":"Python","stars":1159,"forks":299,"topics":["aiops","gemini","hacktoberfest","haystack","langchain","langraph","llamaindex","llmops","llms","mcp","openai","openai-agents","opentelemetry","pydantic-ai","smolagents","telemetry","tracing","vercel","vertex"],"archived":false,"github_pushed_at":"2026-08-20T18:24:56+00:00","maintenance_label":"Very active","stars_delta_30d":55,"url":"https://www.graphcanon.com/tools/arize-ai-openinference","markdown_url":"https://www.graphcanon.com/tools/arize-ai-openinference.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/arize-ai-openinference","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=arize-ai-openinference","description":"OpenTelemetry Instrumentation for AI Observability","homepage_url":"https://arize-ai.github.io/openinference/","license":"Apache-2.0","open_issues":238,"watchers":15,"ai_summary":"The OpenInference project provides OpenTelemetry instrumentation tools and specification to improve observability in the context of AI applications, including LLMs and AI agents.","readme_excerpt":"

\n \"OpenInference\"\n

\n

\n \n \n \n \n \n \n

\n\nOpenInference is a set of conventions and plugins that is complimentary to [OpenTelemetry](https://opentelemetry.io/) to\nenable tracing of AI applications. OpenInference is natively supported\nby [Arize Phoenix](https://github.com/Arize-ai/phoenix) and [Arize AX](https://arize.com/products/ax?utm_source=docs&utm_medium=web&utm_content=openinference), but can be used with any OpenTelemetry-compatible backend as\nwell.\n\n## Specification\n\nThe OpenInference specification is edited in markdown files found in the [spec directory](./spec/). It's designed to\nprovide insight into the invocation of LLMs and the surrounding application context such as retrieval from vector stores\nand the usage of external tools such as search engines or APIs. The specification is transport and file-format agnostic,\nand is intended to be used in conjunction with other specifications such","github_created_at":"2023-12-26T17:33:58+00:00","created_at":"2026-07-07T17:43:04.372698+00:00","updated_at":"2026-08-21T00:00:57.028321+00:00","categories":[{"slug":"evaluation-observability","name":"Evaluation & Observability","url":"https://www.graphcanon.com/categories/evaluation-observability","markdown_url":"https://www.graphcanon.com/categories/evaluation-observability.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/evaluation-observability"}],"tags":[{"slug":"aiops","name":"aiops"},{"slug":"llmops","name":"llmops"},{"slug":"openinference","name":"openinference"},{"slug":"telemetry","name":"telemetry"},{"slug":"tracing","name":"tracing"}],"trust":{"provenance":{"is_fork":false,"github_id":735999703,"owner_type":"Organization","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-08-21T00:00:56.253Z","maintenance":{"label":"Very active","score":96,"methodology":"github_public_v1","releases_90d":30,"days_since_push":0,"last_release_at":"2026-08-18T07:46:44Z","stars_delta_30d":55,"open_issues_delta_30d":5},"security_summary":{"status":"no_manifest","scanner":null,"low_count":0,"high_count":0,"last_scan_at":"2026-07-11T11:22:45.191Z","medium_count":0,"scan_profile":"mcp_manifest","critical_count":0}},"capability_facts":{"scan":{"source":"repo_scan","observed_at":"2026-08-21T00:00:56.714Z"},"languages":{"value":["python"],"source":"github.language","observed_at":"2026-08-21T00:00:56.714Z"},"license_spdx":{"value":"Apache-2.0","source":"github.license","observed_at":"2026-08-21T00:00:56.714Z"}},"decision_facts":{"hosting":null,"pricing":null,"requirements":null,"constraints":null,"when_to_use":["When you are working with AI applications that need detailed tracing capabilities alongside observability features provided by OpenTelemetry.","If your project already includes integration or a dependency on Arize tools like Arize Phoenix or Arize AX, making this tool natively compatible for enhanced functionality."],"when_not_to_use":["When your AI applications do not require the specific tracing features supported by OpenInference and can operate effectively with standard OpenTelemetry capabilities alone.","If you are looking for alternatives that offer more generic observability features without needing to integrate Arize-specific tools, as this might make OpenInference less beneficial."],"source":"enrich:decision_facts","observed_at":"2026-07-09T09:08:16.028Z"},"constraint_facets":null,"decision_summary":[{"label":"Adopt for","value":"OpenInference is a tool focused on providing observability for AI systems using OpenTelemetry, designed to complement it by enabling tracing of AI applications. It is natively supported by Arize Phoenix and AX but can be"}]}}